Speaker Diarization 2.5

🚀 Speaker diarization 2.5

This is a speaker diarization pipeline modified from pyannote/speaker-diarization-3.0, offering efficient and accurate speaker segmentation and identification.

🚀 Quick Start

Using this open-source model in production?
Consider switching to pyannoteAI for better and faster options.

This pipeline is modified from pyannote/speaker-diarization-3.0. It uses pyannote/segmentation-3.0 for Speaker Segments and the Speaker Embedding: speechbrain/spkrec-ecapa-voxceleb from pyannote/speaker-diarization@2.1. For some testings, Embeddings from speechbrain/spkrec-ecapa-voxceleb seems better to detect automatic number of speakers.

✨ Features

• Speaker Segmentation: Utilizes pyannote/segmentation-3.0 for accurate speaker segments.
• Speaker Embedding: Employs speechbrain/spkrec-ecapa-voxceleb for better speaker identification.
• Automatic Speaker Detection: Can automatically detect the number of speakers in some cases.

📦 Installation

1. Install pyannote.audio 3.0 with pip install pyannote.audio
2. Accept pyannote/segmentation-3.0 user conditions
3. Create access token at hf.co/settings/tokens.

💻 Usage Examples

Basic Usage

# instantiate the pipeline
from pyannote.audio import Pipeline
pipeline = Pipeline.from_pretrained(
  "Willy030125/speaker-diarization-2.5",
  use_auth_token="HUGGINGFACE_ACCESS_TOKEN_GOES_HERE")

# run the pipeline on an audio file
diarization = pipeline("audio.wav")

# dump the diarization output to disk using RTTM format
with open("audio.rttm", "w") as rttm:
    diarization.write_rttm(rttm)

Advanced Usage

Processing on GPU

pyannote.audio pipelines run on CPU by default. You can send them to GPU with the following lines:

import torch
pipeline.to(torch.device("cuda"))

Real-time factor is around 2.5% using one Nvidia Tesla V100 SXM2 GPU (for the neural inference part) and one Intel Cascade Lake 6248 CPU (for the clustering part). In other words, it takes approximately 1.5 minutes to process a one hour conversation.

Processing from memory

Pre-loading audio files in memory may result in faster processing:

waveform, sample_rate = torchaudio.load("audio.wav")
diarization = pipeline({"waveform": waveform, "sample_rate": sample_rate})

Monitoring progress

Hooks are available to monitor the progress of the pipeline:

from pyannote.audio.pipelines.utils.hook import ProgressHook
with ProgressHook() as hook:
    diarization = pipeline("audio.wav", hook=hook)

Controlling the number of speakers

In case the number of speakers is known in advance, one can use the num_speakers option:

diarization = pipeline("audio.wav", num_speakers=2)

One can also provide lower and/or upper bounds on the number of speakers using min_speakers and max_speakers options:

diarization = pipeline("audio.wav", min_speakers=2, max_speakers=5)

📚 Documentation

This pipeline has been benchmarked on a large collection of datasets. Processing is fully automatic:

• no manual voice activity detection (as is sometimes the case in the literature)
• no manual number of speakers (though it is possible to provide it to the pipeline)
• no fine-tuning of the internal models nor tuning of the pipeline hyper-parameters to each dataset

... with the least forgiving diarization error rate (DER) setup (named "Full" in this paper):

• no forgiveness collar
• evaluation of overlapped speech

📄 License

This project is licensed under the MIT license.

⚠️ Important Note

The collected information will help acquire a better knowledge of pyannote.audio userbase and help its maintainers improve it further. Though this pipeline uses MIT license and will always remain open - source, we will occasionnally email you about premium pipelines and paid services around pyannote.

📚 Citations

@inproceedings{Plaquet23,
  author={Alexis Plaquet and Hervé Bredin},
  title={{Powerset multi-class cross entropy loss for neural speaker diarization}},
  year=2023,
  booktitle={Proc. INTERSPEECH 2023},
}

@inproceedings{Bredin23,
  author={Hervé Bredin},
  title={{pyannote.audio 2.1 speaker diarization pipeline: principle, benchmark, and recipe}},
  year=2023,
  booktitle={Proc. INTERSPEECH 2023},
}